Testosterone What to Measure & When Godfrey C Moses, BSc, MSc,, PhD, FCACB Laboratory Director/Discipline Leader Chemistry, Gamma-Dynacare Medical Laboratories, London, ON (mosesg@gamma( mosesg@gamma-dynacare.com)
General Outline My Interest in Testosterone Measurements Brief Review of Testosterone & Clinical Significance Some Factors Affecting Interpretation of Testosterone Results Technical Issues Associated with Total Testosterone Measurements What & When to Measure Recommendations from 2 major studies Summary & Acknowledgements
Background Why no stated age-specific RI s s for Serum Free Testo in Women when most Labs & Textbooks did; plus such RI s s were vital in diagnosing & treating his patients? Why < 12 pmol/l as RI for FT in women when a result of 0 pmol/l is pathological for all post-pubertal pubertal ages?.. comprehensive review of RI s s Free & Total Testo and participation in the OSCC Consensus Conference on Functional Testo Measurements in Males & Females.
What is testosterone & why measure it? C19 steroid
Multiple Forms Tissue form (Dihydro( Dihydro-Testo DHT) Circulatory form (bound & unbound to proteins) Major androgen; Stimulate spermatogenesis Synthesis & biochemical function - under Hypothal- Pit Control
Clinical Significance Increased in: Precocious Puberty Androgen resistance Ovarian cancer Congenital Adrenal Hyperplasia Polycystic Ovary Disease Testicular Cancer Decreased in: Hypopituitarism Testicular Failure Delayed puberty Chronic illness Prolactinoma
Delta 5 Pregnenolone is an Obligatory Intermediate in its Biosynthesis..
What is the Current Status? Routinely assayed in Serum or Plasma as.. Total Testosterone (TT) nmol/l Unbound or Free Testosterone (FT) pmol/l(2-3 3 % of TT) Bioavailable Testosterone (BAT) nmol/l(43% of TT) Free Androgen Index (FAI; TT/SHBG) Units: 1, 10, 100 SHBG nmol/l(55% of TT) DHT (Tissue Form) Not Done Routinely
Factors Affecting Interpretation RI s Reviewed Published RI s s for FT in Females 11 major providers, US/Canada 7 Reference Labs 1 Teaching Hospital Lab 2 Text Books 1 Manufacturer Kit Insert Analog 5 state age-specific RI s s in adults only 1 states RI in children age 1 10 yrs (not-age specific) Low end of RI ranged from 0.1 3.1 pmol/l 1 states low end RI as less than Ref: Moses, G.C (2001) Unpublished Data
RI - Quebec (Sherebrooke) Study CHUS on Young Healthy Women (N=188, 18 40Y) Test ULR TT 2.54 nmol/l (cf( 2.7 nmol/l) BAT 0.90 nmol/l (cf( 1.2 nmol/l) FT (Calc) 43.4 pmol/l (cf( 58.0 pmol/l) SHBG UTD (because study incl women on the pill) Ref: Fink, G.D (2007) Clin.Chem.2007; 53(6)
Actual Patient Results (LIS Data) show 6 % of men & 17% of women had undetectable TT (<0.3nmol/L) 2 months period (N=5714) Males: 3286 Females: 2438 TT % Males Females <0.3 nmol/l 6.0 17.4 <12.0 nmol/l 4.4 99.4 >10.0 nmol/l 50.5 0.7 <3.0 nmol/l 12.0 96.6 Ref: Moses, G.C (2004 Unpublished Data)
Factors Affecting Interpretation Factors Affecting Interpretation.... Biological Variation
Factors Affecting Interpretation Factors Affecting Interpretation. Biological Variation
Factors Affecting SHBG Increased in Estrogen Treatment, Androgen and Growth Hormone deficiencies, Aging, Thyrotoxicosis, Alcoholic Cirrhosis and Hepatitis Decreased in Hyperinsulinemia,, Obesity, Acromegaly,, Androgen Treatment, Hypothyroidism, Nephrotic Syndrome, Hypercortisolemia,, Familial Disorders
How is TT measured & what are some of the issues? TT: Immunoassays Direct & Indirect (Extraction); Direct Automated Preferred; Chromatography HPLC & GC Mass Spectrometry; Tedious/unsuited for routine use FT: Measured - Immunoassays RIA & Analoq; Equilibrium Dialysis Calculated (1) {TT [AlbT+SHBGT]}; (2) [BAT AlbT] BAT: Measured - Immunoassays RIA/AS ppt Calculated (1) {FT +AlbT+ AlbT}; (2) [TT SHBGT] ED is gold standard but not suitable for routine use Analoq-based FT has no added clinical information to measured TT (Fritz, K. S. et al. Clin Chem 2008; 54: 512 516) Majority of users in CAP Survey still use the analog-based FT
Few Labs Perform SHBG & FAI FAI (Free Androgen Index) & SHBG: FAI = TT nmol/l / SHBG nmol/l x 10, x100, x1000 Result is method dependent SHBG immunoassay; manual; few automated assays exists but not available in most labs Calculated FT, BAT & FAI must have accurate & precise SHBG assays AS ppt method of choice BAT but results can vary from lab to lab (technique)
Issues with assays for multiple components of Testo IA s: Perform differently in males and females Cross reactivity with other steroids Lack Adequate Anal Sens for use in kids, women & elderly men with diagnosed androgen deficiency Others: Lack of evidence-based protocol for BAT Age-associated discrepancy between mbat and cbat in men; a decrease in cbat is aprrox. 2 x mbat (Dechaud,, H. et al. Clin Chem 2007; 53(4): 723 728). When to do further work-up? On females if TT by automated IA s s >1.2nmol/L, >3nmol/L or 5 nmol/l? (> 3 nmol/l most common); SHBG, extract in- house and/or send out to another lab with different IA method or IDMS reference lab?
IA s for TT perform differently in men & women.. Immunoassays vs GC-MS Total Testo; ; n=10 IA vs ID/GCMS 7/10 had higher values than GCMS (up to 46% higher) in Females 4/10 had lower values than GCMS (up to 46% lower in Males Magnitude of Mean Differences (Bias) were also different in males & females 0/10 was sufficiently reliable for use in Children & women 7/10 had r > 0.95 but only in men; r < 0.95 for females in all 10. Ref. 1. Taieb et al. et al. Clin.. Chem. 2003; 49(8): 1381-1391. 1391.
IA s for TT perform differently in men & women.. Female TT GCMS Targeting Exercise. Sept 2002; N=218 labs All had Slope > 1.00 relative to GCMS, except Bayer Centaur, Beckman Access and an in-house RIA extraction method. 75% had + bias in normal range & 50% had + bias at traditional ULN Between and within method variability was Huge Ref. Jonathan Middle. UK NEQAS Steroid Hormone 2002 Annual Review
Several Formulae for calculating FT & BAT are available, but need to keep in mind.. Calculations are based on binding constants at physiological protein concentrations Provide results in either SI or non-si units, expressed as absolute values or percentages of TT Calculation of BAT & FT in men: A Comparison of 5 Published Algorithms (De( Ronde,, W. et al. Clin Chem 52:177-1784 1784) Chem 2006; The International Society for the Study of the Aging Male (ISSAM) Website (www.issam.ch/freetesto.htm( www.issam.ch/freetesto.htm)
Performance Goals lack of standardized PG s Total Error Concept (CV anal < 0.5CV intra & Bias < 0.25 (CV intra + CV inter) FT TT Desirable (%) 24.4 13.0 13.0 31.6 SHBG BAT 31.6 - Optimal (%) 16.3 6.5 21.1 - Anal CV (%) 7.2/16.6 8.7/7.9 12.4/11.2 - TT: Abbott Architect; FT: DPC Analogue Direct; SHBG: Manual RIA
Reference Change Value (RCV,%) Utility FT TT SHBG BAT L1 L2 L1 L2 L1 L2 L1 L2 95 % CL 52.7 45.7 34.3 32.8 48.0 32.6 - - 99% CL 68.9 42.6 44.8 42.8 62.7 59.7 - - [Anal & Intra-Individual Individual Variation: {1.414 x Z x SQRT (CV 2 anal + CV 2 intra )}] x100]
There is not always Gold in the Gold Standard - Ref: Joint Committee for Traceability in Lab Medicine Database (www.bipm.org/jctlm) IDMS Methods for Testosterone in Blood Serum. Method 1 DGKC definitive method for serum T Method description ID/GC/MS Applicable Range 2 nmol/l to 45 nmol/l Expected Uncertainty (95% CL) 1 % to 3 % Reference(s) Siekmann et al., J. Steroid Biochem., 1979, 11, 117-123 123 Method 2 Univ. of Ghent Rf.. Meas. Proced.. for T Method description ID/GC/MS Applicable Range 0.05 nmol/l to 70 nmol/l Expected Uncertainty (95% CL) 1.5 % to 3 % Reference(s) Anal Chem., 199, 66, 4116-41 Eur.. J. Clin.. Chem. Clin. Biochem., 1996, 34, 853-860 860 LC-MS/MS AACC: Abstr # 189. Anal Sens: : 0.07 nmol/l; AMR: 0-129 nmol
How is Clinical Applicability of GC- & LC-MS Techniques Assessed? Comparison with Direct Automated Immunoassays Clinical Data (evidence-based medicine) Needed Varying Analytical Sensitivities Detection Limits 10 Immunoassay Systems (TT; nmol/l) Assay System Detection Limit (manufacturer s s claim) Abbott Architect 0.28 Bayer Centaur 0.35 Bayer Immuno1 0.17 Bio-Merieux (Vidas) 0.35 Immunlite 2000(DPC) 0.30 Vitros Eci (OCD) 0.03 Perkin Elmer AutoDelphia 0.40 Roche Elecsys 2010 0.069 Beckman (ImmunoTech( ImmunoTech) 0.086 Dade Behring (PC) 0.14 Ref. 1. Taieb et al. Clin Chem 2003; 49(8): 1381 1391.
What & When to Measure Testosterone & Which Method/Assay?
Ontario Community Labs Perform More FT Than TT Annually Total Free BAT SHBG 2003 2004 Hospital Labs 25565 9869 - - Community labs 95681 100284 - - Lab Licensing Testing Labs Data ( 04)*( Hospital Labs 325** 1200 Community Labs 12800 420 *Data is annualized and do not include all provincial Labs; ** some s regional hosp. labs do not perform Total T or BAT, while others are looking at offering a calculated BAT in the near future.
What is really driving the continued interest in Testing for Testosterone? Need for prompt diagnosis and treatment of hypogonadism in elderly men Testosterone (sex hormone) concentration and self-rated health & life satisfaction, depression or dementia in the elderly Improved Sex Drive/Sexual Performance in Men Increased Libido and Mood in women Testosterone (sex hormone) influence on pubertal growth
What is driving interest cont d Erectile Dysfunction (ED) - Relatively Common; Affects Men & Women (Dean( Dean J. et al. Int J Clin Clin Practice. 2008; 62: 127-133 133 & Fisher W.A. et al. FEMALES Study Females Experience of Men s s Attitudes to Life Events and Sexuality J Sex Med. 2005; 2: 675-672) Evidence that testosterone treatment is beneficial Improved/enhanced sexual function and mood Decreased body fat mass Increased lean body mass Increased bone mineral density Normalized blood total & free testosterone in deficient subjects Ref. Wang et al. J Clin Endocrinol Metab.. 2004; 89:2085 2098 & Synder et al. J Clin Endocrinol Metab.. 2000; 85:2670 2677)
What is driving interest cont d T Replacement Therapy In US # prescriptions for T replacement increased by 1700% from 1994 2003 (Bhasin,, S. et al. J Andol.. 2003; 24: 299-311.. recommendations for monitoring & managing risks of prostate disease during T replacement therapy in older men) Finasteride Tx for BPH (F is 5-alpha 5 reductase inhibitor) effects on T replacement? In Finland, nearly 100% increases in wholesale turnover for T between 1993 & 2004 (E472.9K to E913.4K)
Evidence for multiple health & functional benefits for testosterone supplementation, particularly often sought by the elderly, real? Finish Study Associations of Sex Hormone Concentrations with Health and Life Satisfaction in Elderly Men (Eskelinen,, S. I. et al. Endocr Pract. 2007; 13(7): 743-749. 749. Also on www.medscape.com/viewarticle/572248) 1998-1999, N = 517 elderly men (mean age, 73 years; range, 64 to 97) Serum levels of TT, LH, follicle-stimulating hormone (FSH), sex hormone-binding globulin (SHBG), and estradiol (E ) were 2 determined by the AutoDelfia automatic immunoassay (Time- resolved fluorescent immunoassay). Non-SHBG SHBG-bound T (cft( cft/bat) calculated as follows: %FT = 2.28-1.38 log(shbg [nmol/l]/10)
Evidence of benefits.. Finish Study Cont d Serum FT (pmol( pmol/l) = %FT TT (nmol( nmol/l) 10. The analytical sensitivity of the TT assay was 0.3 nmol/l. Calibrated with IDMS traceable standard. Self-rated Health, Life Satisfaction, Self-Rating Depression & Dementia (21 Q questionnaire) Conclusion Low serum T concentration was associated with diagnosed depression. Subclinical hypogonadism may be associated with diagnosed dementia. Older age and being overweight were predictors of poor health or occurrence of many symptoms among elderly men.
What to Measure & When? OSCC Experts OSCC Consensus Conference C. P. COLLIER, et al. Functional testosterone: Biochemical assessment of hypogonadism in men Report from a multidisciplinary workshop hosted by the Ontario Society of Clinical Chemists. The Aging Male 2008; 10(4): 211 216 Objectives: to establish the most reliable, cost-effective approach to the biochemical assessment of hypogonadism in men. Who: Specialists across Canada in clinical biochemistry, endocrinology, family medicine and urology (1-day workshop)
OSCC Experts Con t Method: Two key challenges - 1) Total testosterone (TT) determinations are equivalent to bioavailable testosterone (BAT) or calculated BAT (cbat( cbat) ) or free testosterone (FT) (by analogue radioimmunoassay or equilibrium dialysis) or calculated FT (cft( cft); 2) No good evidence that borderline low testosterone concentrations in men should be treated. Outcomes: Identify what agreement exists in Canada, what issues were still controversial, and what research remains to be addressed Conclusions. Total testosterone is a reliable marker for the initial investigation of men presenting with symptoms of hypogonadism; cbat is a reasonable follow-up test in patients with equivocal biochemical or consistent symptomatic findings.
What to Measure & When? OSCC Expert Panel Recommendations.. Start with AM (before 11:00 am) total testosterone (TT) measurement TT < 8nmol/L & high probability of hypogonadism,, repeat TT with baseline LH & prolactin TT > 15nmol/L; hypogonadism not likely; however, if symptoms persist, repeat TT & baseline LH & prolactin TT = 8 15 nmol/l, repeat TT & automatically reflex to SHBG if repeat is still 8 15 nmol/l and cbat (lab should provide cbat automatically); Baseline LH and prolactin as well
OSCC Expert Panel Recommendations Cont d Measured FT by analogue should Not be used in diagnosis of hypogonadism; ; Should be replaced by cbat using SHBG & TT (note: did not favour mbat) Provincial governments should support the development of IDMS reference testing laboratory to assure the accuracy and reliability of testosterone assays. - Large study in elderly men is needed to fully addressed challenge #2
Summary: What to Measure & When Androgen excess in males & females, at any age - TT by conventional IA In females, be mindful of falsely elevated results; Access to alternate direct auto IA, pre-treatment (extraction) with rerun, FAI/FT or BAT Hypogonadal function in males TT by conventional IA in most cases If sex hormone binding globulin abnormality is suspected in hypogonadic states, BAT and/or measured or calculated FT In pre-pubertal pubertal children (males & females) and normal females TT by HPLC or Mass Spec; need to established RI s
Summary: What to Measure & When Cont d. Treatment and monitoring clinically diagnosed hypogonadism TT by conventional IA; if TT < 5 or > 10 nmol/l, no other test If TT 5 10 nmol/l, repeat; If still between 5 10 nmol/l after repeat; measure LH and calculated or measured BAT AM testosterone (before 11.30 AM) Testosterone supplementation for health & function benefits in older men (and women?) without symptoms of androgen deficiency is still controversial but TT by IA is adequate Testosterone replacement in older men with symptoms suggestive of androgen deficiency TT by IA, LH and FT, if necessary.
Acknowledgements Drs Chris Collier & Hilde Vandenberge Technical and IT Staff at GDML OSCC Executive Colleagues at CML & LifeLabs